Hi all,

This is my first post but i'll try to make my questions as clear as possible. To be clear this is my first time using Stata.

1) When i use ivregress 2sls it reports z scores and z p-values (same when using ivreg2). Using just ivreg with multiple instruments reports the same coefficient, different standard errors and t-statistics instead. My project follows the work of an academic paper and i'm recreating some of the tables from it with my own data. They only report t-statistics and my understanding was that unless we know the parameters of the population we should use t-statistics not z scores. I've searched the web looking for help: One site suggests the reported z-values are identical to the t-statistics reported when the same regressions are ran in a different software - and so the z-value in stata is just the t-stat with a different label? If anyone has any insights on this i'd be greatly appreciative.
[ATTACH=CONFIG]temp_21666_1615812656355_806[/ATTACH]

2) I'm having trouble correctly recreating a bar chart. I've attached an image of what it's supposed to look like, what mine currently looks like and the Stata code used to generate it.
[ATTACH=CONFIG]temp_21667_1615812781544_563[/ATTACH]
Array



Code:
collapse (mean) bank lly privy gdppc_usd05, by(countryid)
xtile quart_income = gdppc_usd05, nq(4)
graph bar (mean) lly (mean) bank (mean) privy, over(quart_income)
The chart is supposed to show that generally the indicators for financial development (lly, privy, bank) increase as we move from low to high income countries. I collapsed the panel data to average each variable for each country over the 30yr period and generated quartiles of income. Ideally the legend would contain the income group quartiles, and the groupings be for each independent variable. As you can see my chart is the reverse.

I do apologize if this is a too basic question but after watching multiple youtube guides and reading through the manual, I still cannot seem to figure it out.

Thank you for any help and guidance,

John